A randomized controlled study evaluated an Internet-based HIV behavioral intervention for men who have sex with men (MSM). Hong Kong Chinese men aged >or=18 having engaged in oral or anal MSM behaviors in the last six months were randomly allocated into the intervention (n=140) and the control groups (n=140). Participants were interviewed anonymously at the baseline and after six months. Interventions included periodic HIV information dissemination, monitoring of risk behaviors and interactive feedback, online peer counseling and provision of a hotline. The control group only received some educational materials. Efficacy of the intervention was not supported as no statistically significant between-group differences in background characteristics, HIV risk behaviors and perceptions were observed both at the baseline and after the 6-month period. Effectiveness of online interventions should not be taken for granted and evidence-based approach is warranted.